Search NASA⌕ Search

SEARCH · Search NASA

Results for “continuous time system estimation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 127 records · Page 7

A Data-Driven Nonparametric Approach for Probabilistic Load-Margin Assessment Considering Wind Power Penetration

A modern power system is characterized by an increasing penetration of wind power, which results in large uncertainties in its states. These uncertainties must be quantified properly; otherwise, the system security may be threatened. Facing this challenge, here we propose a cost-effective, data-driven approach to assessing a power system's load margin probabilistically. Using actual wind data, a kernel density estimator is applied to infer the nonparametric wind speed distributions, which are further merged into the framework of a vine copula. The latter enables us to simulate complex multivariate and highly dependent model inputs with a variety of bivariate copulae that precisely represent the tail dependence in the correlated samples. Furthermore, to reduce the prohibitive computational time of traditional Monte-Carlo simulations that process a large amount of samples, we propose to use a nonparametric, Gaussian-process-emulator-based reduced-order model to replace the original complicated continuation power-flow model through a Bayesian-learning framework. To accelerate the convergence rate of this Bayesian algorithm, a truncated polynomial chaos surrogate, which serves as a highly efficient, parametric Bayesian prior, is developed. This emulator allows us to execute the time-consuming continuation power-flow solver at the sampled values with a negligible computational cost. Results of simulations that are performed on several test systems reveal the impressive performance of the proposed method in the probabilistic load-margin assessment.

17 WIND ENERGY↗

New detector for use in fast neutron radiography

We have developed and tested a new type detector for use in the fast neutron (FN) imaging radiography applications. FN radiography is generally used for nondestructive material testing, medical and biology applications, border patrol, transportation and cargo screening tasks. It is complementary to other types of radiography, providing additional information on light element content of the material samples. Distinct from other FN imagers presently known, our device implements a neutron-sensitive scintillator attached to a position-sensitive photomultiplier tube (PSPMT), and operates in an event-by-event readout mode, acquiring energy, timing, and pulse shape information for all detected radiation events. The information is used to help separate events of FN interactions in the scintillator from the background events caused by the electronics noise and by the other types of background radiation. Selection of pure fast neutron events in the final image allows us to achieve ultimate image contrast and resolution, as compared with other types of FN imaging devices operating most commonly in an integration mode, in which the detector?s dark noise and radiation background dilute the images. The detector performance for FN imaging application was tested using D-D neutron generator, designed and manufactured by Adelphi Technology, Inc. This essentially point-like neutron source operates in continuous mode producing up to 109 of 2.5 MeV neutrons per second. Samples made of metals plastic and other material were used to measure the detector resolution, efficiency and uniformity. Results of these tests are presented and discussed. Fig. 1 shows one of the test FN radiographic images obtained using the sample made of 11 styrene plastic strips. All strips are squares 4.8 x 4.8 mm2 with six different lengths 10 to 60 mm with 10 mm increment. [A] [B] [C] Fig. 1. [A]-layout of the test sample; [B]-raw FN shadow image of the sample; [C]-map of the plastic strips as they appear on the image, with measured FN absorption factor shown in percent. Both X and Y position resolution of the FN imaging detector are estimated to be less than 0.5 mm (sigma) based on the analysis of the edges of the strip shadows in the plot. Because this detector shows the fraction-of-a-millimeter resolution desirable for most of FN applications, and is built using radiation hard materials, we believe it could be a good alternative to other FN imaging systems based on CCD or solid state detectors. In addition, having a sub-nanosecond timing resolution, it is suitable for the time-of-flight energy-resolved FN imaging.

Popov, V.↗

On fast simulation of dynamical system with neural vector enhanced numerical solver

The large-scale simulation of dynamical systems is critical in numerous scientific and engineering disciplines. However, traditional numerical solvers are limited by the choice of step sizes when estimating integration, resulting in a trade-off between accuracy and computational efficiency. To address this challenge, we introduce a deep learning-based corrector called Neural Vector (NeurVec), which can compensate for integration errors and enable larger time step sizes in simulations. Our extensive experiments on a variety of complex dynamical system benchmarks demonstrate that NeurVec exhibits remarkable generalization capability on a continuous phase space, even when trained using limited and discrete data. NeurVec significantly accelerates traditional solvers, achieving speeds tens to hundreds of times faster while maintaining high levels of accuracy and stability. Moreover, NeurVec’s simple-yet-effective design, combined with its ease of implementation, has the potential to establish a new paradigm for fast-solving differential equations based on deep learning.

97 MATHEMATICS AND COMPUTING↗

Model Calibration of Exciter and PSS Using Extended Kalman Filter

Power system modeling and controls continue to become more complex with the advent of smart grid technologies and large-scale deployment of renewable energy resources. As demonstrated in recent studies, inaccurate system models could lead to large-scale blackouts, thereby motivating the need for model calibration. Current methods of model calibration rely on manual tuning based on engineering experience, are time consuming and could yield inaccurate parameter estimates. In this paper, the Extended Kalman Filter (EKF) is used as a tool to calibrate exciter and Power System Stabilizer (PSS) models of a particular type of machine in the Western Electricity Coordinating Council (WECC). The EKF-based parameter estimation is a recursive prediction-correction process which uses the mismatch between simulation and measurement to adjust the model parameters at every time step. Numerical simulations using actual field test data demonstrate the effectiveness of the proposed approach in calibrating the parameters.

Kalsi, Karanjit↗

2016 Alabama PV soft cost and workforce development

The Southeastern US has the largest potential for growth in the solar industry. However, currently they languish behind the rest of the US. There are several bright spots including the large number of utility scale installations in North Carolina and the recent successes in South Carolina under Act 236. In order to better understand the impacts of state legislation on the growth of the solar industry in the SE US, the Savannah River National Laboratory has undertaken a study to look at the growth in each state in order to develop recommendations to help reduce the cost of solar and to spur the industry. This is the second report in the series. The first focused on developing cost metrics for South Carolina under Act 236. This report focuses on Alabama, the 49th ranked state for solar business, which has very similar population and median income to South Carolina. For this survey, the ten known in-state installers were contacted. Responses were received from seven, representing 70% of the installers, a majority of which provide both residential and commercial installations. Interestingly, none of the respondents serve the utility scale sector. Overall, costs for Alabama are on track with the rest of the country with a reported average cost of $3.29/W-DC for residential systems and $2.44/W-DC for commercial systems. 60% of this cost is attributed to hardware only. Of the remaining costs, installation contributed to the largest percentage of soft costs followed by overhead, marketing and sales, and permitting, respectively. This also closely mirrors results seen in South Carolina. Job growth in the industry is expected to proceed well. An expected 34-42 additional full time equivalent jobs were expected to be added in Alabama within the six month window following the survey period. During the three years following the survey, this number was expected to double with 89-97 additional jobs being added to the market. In both cases, a vast majority of these jobs were for installation professionals and electricians. Despite the cost of solar, the industry continues to struggle in Alabama, largely due to the absence of any statewide net metering legislation. By current best estimates, there are over 60 residential installations statewide; however, this number is difficult to track due to the lack of a State authority keeping a consolidated list of grid connected distributed power systems. In South Carolina, the Energy Office tracks and reports grid connected distributed power systems by all Cooperatives and utilities. In Alabama, the Energy Office does not fulfill this role and data must be collected directly from each utility and cooperative, which makes collection and analysis difficult. Having a central state agency track this information would be extremely useful towards developing state policy recommendations, particularly if net metering were enabled within the state.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Advanced, Energy-Efficient Hybrid Membrane System for Industrial Water Reuse

In the U.S. manufacturing sector, current industrial water use practices are energy-intensive and utilize and discharge high volumes of waters, rendering them not sustainable especially in light of the growing scarcity of suitable water supplies. To help address this problem, the goal of this project was to develop an advanced, cost-effective, hybrid membrane-based water treatment system that can improve the energy efficiency of industrial wastewater treatment while allowing at least 50% water reuse efficiency. This hybrid process would combine emerging Forward Osmosis (FO) and Membrane Distillation (MD) technology components into an integrated FO-MD system that can beneficially utilize low-grade waste heat (i.e., T < 450 °F) in industrial facilities to produce distilled-quality product water for reuse. In this project, laboratory-, bench-, and pilot-scale experiments on the hybrid FO-MD system were conducted for industrial wastewater treatment. It was demonstrated at laboratory, bench, and pilot scales that FO-MD membrane technology can concentrate brine to very high total dissolved solids (TDS) levels (>200,000 ppm) that are at least 2.5 times higher than the TDS level to which RO can achieve. In laboratory testing, currently available FO and MD membranes were tested to select for high-performing membranes with high salt rejection and high water flux. Multiple FO membrane/draw-salt solution combinations that gave high water flux with higher than 98% salt rejection were also identified. Reverse draw-salt fluxes were observed to be much lower for divalent salts than for monovalent salts. MD membranes were identified that had 99.9+% salt rejection and water flux as high as 50-90 L/(m 2 ·h) for flat-sheet membranes and >20 L/(m 2 ·h) for hollow fibers. In bench-scale testing, a single unit of commercially available FO and MD membrane modules were evaluated for continuous, integrated operation. Using the laboratory- and bench-scale test data, numerical modeling was performed on the FO and MD processes to estimate engineering parameters for a larger-scale pilot unit. Based on the experimental studies and modeling results, a pilot-scale, integrated FO-MD prototype unit was designed and built for trailer-mounted operation. This prototype system was fed real industrial wastewater, which could not be further treated by conventional technologies, from an oil production facility and was successfully operated for over 15 weeks without major stoppage. About 90% water recovery was possible, while concentrating the TDS from 12,000 ppm up to 190,500 ppm. The FO-MD prototype rejected most wastewater contaminants while producing water with <300 ppm TDS, even when the feed TDS was higher than 150,000 ppm. No chemical cleaning was necessary during the pilot testing period. Flushing the system with dechlorinated tap water was sufficient to reset the membranes for the next set of test conditions. Pilot performance and membrane autopsy showed that, even though the feed was concentrated more than 10 times, membrane fouling was unnoticeable and no defects were detected on the FO and MD membrane surfaces. This project demonstrated the technical feasibility of the hybrid FO-MD process by taking water already treated to the limit with the highest level of current technologies and further concentrating it 10-fold by using mostly low-cost materials. Because no membranes suitable for full-scale plant applications are available at present, economical feasibility of the hybrid technology is still uncertain, but it is expected that broader industry participation can further reduce FO-MD process costs.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

EGS Collab Experiment 1: Microseismic Monitoring

The U.S. Department of Energy's Enhanced Geothermal System (EGS) Collab project aims to improve our understanding of hydraulic stimulations in crystalline rock for enhanced geothermal energy production through execution of intensely monitored meso-scale experiments. The first experiment is being performed at the 4850 ft level of the Sanford Underground Research Facility (SURF), approximately 1.5 km below the surface at Lead, South Dakota. Here we report on microseismic monitoring of repeated stimulation experiments and subsequent flow tests between two boreholes in the Poorman Formation. Stimulations were performed at several locations in the designated injection borehole at flow rates from 0.1 to 5 L/min over temporal durations from minutes to hours. Microseismic monitoring was performed using a dense 3D sensor array including two cemented hydrophone strings with 12 sensors at 1.75 m spacing accompanied by 18 3-C accelerometers, deployed in 6 monitoring boreholes, completely surrounding the stimulation region. Continuous records were obtained over a two-month period using a novel dual recording system consisting of a conventional 96 channel exploration seismograph and a high-performance 64 channel digitizer sampling sensors at 4 and 100 kHz respectively. Using a standard STA/LTA triggering algorithm, we detected thousands of microseismic events with recorded energy in a frequency range generally above 3 kHz and up to 40 kHz. The locations of these events are consistent with creation of a hydraulic fracture and additional reactivation of pre-existing structures. Using manual pick refinement and double-difference relocation we are able to track the fracture growth to high precision. We estimate the times and locations of the fracture intersecting a monitoring and the production borehole using microseismic events. They are in excellent agreement with independent measurements using distributed temperature sensing, in-situ strain observations and measurements of conductivity changes. This submission includes a microearthquake catalog, raw event files, a subset of the continuous microseismic monitoring data collected during stimulations and flow test activity on 05/22/2018, 05/23/2018, 05/24/2018, 05/25/2018, 06/25/2018, 07/19/2018, 07/20/2018, 12/7/2018, 12/20/2018, and 12/21/2018 (in binary format), and a binary file interpreter to read the continuous microseismic monitoring data. A Stanford Geothermal Workshop paper is also included to describe microseismic monitoring activities at SURF during these periods.

15 GEOTHERMAL ENERGY↗

A Robust Yule-Walker Method for Online Monitoring of Power System Electromechanical Modes of Oscillation

Reliable power system operation requires that system operators maintain a satisfactory small-signal stability margin at all times. Mode meters allow system operators to continuously monitor the stability margin by analyzing synchrophasor measurements. Experience with mode meters at the Bonneville Power Administration (BPA) has revealed that mode meters sometimes suffer severe bias following transient disturbances. In this paper, a robust version of the Yule-Walker mode meter algorithm is proposed to mitigate this bias. The robust capability is introduced by modifying the estimator of the autocovariance sequence (ACS) on which the algorithm depends. Tests with field-measured data demonstrate that the robust Yule-Walker (RYW) algorithm significantly outperforms its conventional counterpart and is better able to provide system operators with accurate information about the system's stability margin.

Follum, James D.↗

Classification and Localization of Fracture-Hit Events in Low-Frequency Distributed Acoustic Sensing Strain Rate with Convolutional Neural Networks

Summary Distributed acoustic sensing (DAS) has been used in the oil and gas industry as an advanced technology for surveillance and diagnostics. Operators use DAS to monitor hydraulic fracturing activities, examine well stimulation efficacy, and estimate complex fracture system geometries. Particularly, low-frequency DAS can detect geomechanical events such as fracture hits because hydraulic fractures propagate and create strain rate variations in the rock. Analysis of DAS data today is mostly done post-job and subject to interpretation methods. However, the continuous and dense data stream generated live by DAS poses the opportunity for more efficient and accurate real-time data-driven analysis. The objective of this study is to develop a machine learning-based workflow that can identify and locate fracture-hit events in simulated strain rate responses correlated with low-frequency DAS data. In this paper, “fracture hit” refers to a hydraulic fracture originating from a stimulated well intersecting an offset well. We start with building a single fracture propagation model to produce strain rate patterns observed at a hypothetical monitoring well. This model is used to generate two sets of strain rate responses with one set containing fracture-hit events. The labeled synthetic data are then used to train a custom convolutional neural network (CNN) model for identifying the presence of fracture-hit events. The same model is trained again for locating the event with the output layer of the model replaced with linear units. We achieved near-perfect predictions for both event classification and localization. These promising results prove the feasibility of using CNN for real-time event detection from fiber-optic sensing data. Additionally, we use edge detection techniques to recognize fracture-hit event patterns in strain rate images. The fracture-hit location can be identified using recognized pixels in the image. The accuracy of edge detection-based location identification is also plausible, but edge detection is dependent on the assumption of pattern shape and image quality, hence it is less robust compared to CNN models. This comparison further supports the need for CNN applications in image-based real-time fiber-optic sensing event detection.

Engineering↗

Development and Implementation of an Automatic Continuous Online Monitoring and Control Platform for Polymerization Reactions to Sharply Boost Energy and Resource Efficiency in Polymer Manufacturing

The project goal was to create an energy saving paradigm shift in how polymers are manufactured in the 21st century. It used Automatic Continuous Online Monitoring of Polymerization reactions (ACOMP) integrated for the first time with automatic active control to create the innovative ‘ACOMP/Control Interface’, or ‘ACOMP/CI’. ACOMP/CI will begin the transformation from old, inefficient processes into highly evolved, energy and resource efficient ones. The ACOMP platform is broadly applicable to many types of reactions and processes throughout the vast polymer industry. The industry provides materials for sectors such as automotive, aerospace, oil recovery, agriculture, paints, resins, adhesives, pharmaceuticals and therapeutic proteins, optics, electronics, lightweight building materials, and many more. The U.S. chemical industry is one of the last major sectors in which the U.S. has top global stature. It consumes 24% of all U.S. manufacturing energy, produces over $800B of product annually, supports 25% of the U.S. GDP and employs over 6 million people. It is also a major source of GHG emissions. Polymers make up approximately 30% of this sector. It is estimated that annually 60 TBtu of energy could be saved and 3 million tons less of GHG emissions produced by optimizing production in the polyolefin manufacturing sector alone. The project scope included first time design and prototyping of an ACOMP/CI, creation of active reaction controllers, and demonstration of control capabilities on ideal, low concentration polymerization reactions. All these elements of the scope were met, including advances and findings not originally anticipated. Extensions to more complex reactions, beyond the reactor capabilities of the current project ACOMP/CI, such as polyolefins and other high pressure/high temperature reactions, are being proposed in Fall 2017 to CESMII, a DoE based NNMI. The initial proposal was for a three year funded project, but this was reduced to a two year project and budget due to funding constraints. Hence, some of the original plans, such as adaptation of the ACOMP/CI to more relevant industrial processes, such as emulsion and dispersion technologies, could not be carried out. A third year of funding was requested at the end of the project, but DoE did not have resources to grant this. The sub-contractor Fluence Analytics (previously Advanced Polymer Monitoring Technologies, Inc) designed, prototyped, and commissioned a working ACOMP/CI by June 2015. The reaction characteristics to be automatically controlled were i) conversion kinetics, ii) molecular weight, iii) copolymer composition, and iv) simultaneous molecular weight and composition. A two pronged control strategy was used. The Tulane/Fluence group took a basic principles approach that did not rely on kinetic models. The LSU group took a more complex, non-linear model-oriented approach involving complete kinetic descriptions of the reaction system Each of these approaches proved successful in their own way. By April 2016 fully automatic control of conversion and weight average molecular weight, Mw, trajectories was achieved using the Tulane/Fluence (TF) basic principles controller. Similar results were obtained by the LSU non-linear model controller by August 2016. The demonstration system was aqueous free radical polymerization of acrylamide, Am. The control variables were temperature and semi-batch feed to the reactor of Am monomer and initiator. A demonstration of active manual conversion control in an industrial process using high solids in inverse emulsion polymerization of Am was achieved. During Summer 2016 the TF controller was used in conjunction with a chain transfer agent, another control variable, to automatically produce multi-modal molecular weight distributions, MWD, in a single reactor. Industrially, multi-modal MWD are produced by mixing products made in separate reactors, requiring significant extra time, energy, and reactor resources. Recognizing the industrial potential a patent on automatic production of multi-modal polymers was filed, and DoE acknowledged. In Fall 2016 the TF team developed a basic principles controller for copolymer composition and demonstrated it on aqueous free radical copolymerization of comonomers Am and styrene sulfonate, SS. TF then fused the Mw and composition controllers to achieve simultaneous control of both Mw and copolymer composition trajectories. Numerous simultaneous trajectories were demonstrated, including a trimodal composition distribution with constant Mw. Meanwhile, the LSU group developed a Kalman filter to improve the results of their automatic Mw and conversion controller and successful tests were made. During the project the TF team developed a means of computing full MWD during polymer synthesis without need for any chromatographic separation, based on model distributions. This means the polymer product is ‘born characterized’ and this can eliminate post-manufacture analytical laboratory quality control. TF filed a joint patent application on this new approach to chromatography-free determination of MWD with acknowledgment to DoE. The Tulane group obtained a 60MHz NMR during the project and recently completed the first work on separating three comonomers, Am, SS, and Na-acrylate, with a first demonstration of terpolymer composition control with the TF basic principles controller. Widespread dissemination of ACOMP/CI in the polymer manufacturing sector will bolster DoE goals of energy efficiency and reduced GHG emissions: The ability to monitor and actively control polymerization reactions will lead to more efficient use of energy and non-renewable resources, plant and labor time, increase the safety of manufacturing personnel, and will enhance product quality and lead to feasibility of manufacturing of polymers currently too complex for industrial scale production, while leading to less GHG emissions per kilo of product, and allowing for increased U.S. competitiveness in this enormous manufacturing sector. When ACOMP/CI is expanded to the polyolefin industry it is estimated that 60 TeraBTU/year of energy can be saved. Much of this saving is anticipated to come from optimized control of grade changeovers in steady state reactors and maintenance of steady states. Conclusions: ACOMP’s ability to provide continuous realtime data streams of measured polymer and reaction characteristics made it possible, for the first time, to directly and automatically control free radical polymerization reactions. An industrial client of Fluence Analytics has requisitioned the first ACOMP/CI which uses the TF basic principles controller. This sets the stage for FA to add control features to the ACOMP systems it has begun to install on the industrial scale beginning in 2014. Recommendations: This successful project was mainly limited to ideal polymerizations not of an industrial sort. The most energy intensive portion of polymer manufacturing is polyolefins. Adoption of ACOMP/CI to this enormous industrial sector faces the enormous challenges of high temperature, high pressure continuous sampling and high temperature sensor operation to obtain the continuous data needed for direct reaction control. The project team has a strategy for achieving this ambitious goal and will present it in Fall 2017 as a proposal to CESMII/DoE. It is recommended that this upcoming proposal be funded in order to make full use of the achievements of this just ended DoE project as the next step towards making polyolefin ACOMP/CI an energy saving reality. It is projected that ACOMP/CI can have its first polyolefin testbed demonstrations within two years of beginning the proposed project.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Design and implementation of wireless dose logger network for radiological emergency decision support system

A decision support system (DSS) is implemented in Radiological Safety Division, Indira Gandhi Centre for Atomic Research for providing guidance for emergency decision making in case of an inadvertent nuclear accident. Real time gamma dose rate measurement around the stack is used for estimating the radioactive release rate (source term) by using inverse calculation. Wireless gamma dose logging network is designed, implemented, and installed around the Madras Atomic Power Station reactor stack to continuously acquire the environmental gamma dose rate and the details are presented in the paper. The network uses XBee–Pro wireless modules and PSoC controller for wireless interfacing, and the data are logged at the base station. A LabView based program is developed to receive the data, display it on the Google Map, plot the data over the time scale, and register the data in a file to share with DSS software. The DSS at the base station evaluates the real time source term to assess radiation impact.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Developing a Decision Support System for Regional Agricultural Nonpoint Salinity Pollution Management: Application to the San Joaquin River, California

Environmental problems and production losses associated with irrigated agriculture, such as salinity, degradation of receiving waters, such as rivers, and deep percolation of saline water to aquifers, highlight water-quality concerns that require a paradigm shift in resource-management policy. New tools are needed to assist environmental managers in developing sustainable solutions to these problems, given the nonpoint source nature of salt loads to surface water and groundwater from irrigated agriculture. Equity issues arise in distributing responsibility and costs to the generators of this source of pollution. This paper describes an alternative approach to salt regulation and control using the concept of “Real-Time Water Quality management”. The approach relies on a continually updateable WARMF (Watershed Analysis Risk Management Framework) forecasting model to provide daily estimates of salt load assimilative capacity in the San Joaquin River and assessments of compliance with salinity concentration objectives at key monitoring sites on the river. The results of the study showed that the policy combination of well-crafted river salinity objectives by the regulator and the application of an easy-to use and maintain decision support tool by stakeholders have succeeded in minimizing water quality (salinity) exceedances over a 20-year study period.

real-time management economics↗

2025 California Demand Response Potential Study - Charting California’s Demand Response Future. Final Report on Phase 2 Results

California’s legislative and regulatory goals for renewable energy are changing the power grid’s dynamics. Increased variable generation resource penetration connected to the bulk power system, as well as, distributed energy resources (DERs) connected to the distribution system affect the grid’s reliable operation over many different time scales (e.g., days to hours to minutes to seconds). As the state continues this transition, it will require careful planning to ensure resources with the right characteristics are available to meet changing grid management needs. Demand response (DR) has the potential to provide important resources for keeping the electricity grid stable and efficient, to defer upgrades to generation, transmission and distribution systems, and to deliver customer economic benefits. This study estimates the potential size and cost of future DR resources for California’s three investor-owned utilities (IOUs): Pacific Gas and Electric Company (PG&E), Southern California Edison Company (SCE), and San Diego Gas & Electric Company (SDG&E). Our goal is to provide data-driven insights as the California Public Utilities Commission (CPUC) evaluates how to enhance DR’s role in meeting California’s resource planning needs and operational requirements. We address two fundamental questions: 1. What cost-competitive DR service types will meet California’s future grid needs as it moves towards clean energy and advanced infrastructure? 2. What is the size and cost of the expected resource base for the DR service types?

24 POWER TRANSMISSION AND DISTRIBUTION↗

Lyapunov-Based Iterative Learning of Regions of Attraction for Autonomous Systems

This paper proposes a novel algorithm for estimating the region of attraction of equilibrium points for nonlinear discrete-time autonomous systems. The method iteratively expands an initial estimate of the region of attraction by constructing unions of sublevel sets of learned functions parametrized as neural networks. Unlike conventional techniques that rely on a single global Lyapunov function, the proposed approach provides a collection of local Lyapunov-like functions, enabling richer representations and potentially larger region of attraction estimates. These functions are trained using sampled state-space data, and their Lipschitz continuity ensures that desirable properties extend beyond the training samples. The devised strategy is tested via numerical simulations, demonstrating the effectiveness of the proposed approach.

97 MATHEMATICS AND COMPUTING↗

Lyapunov-Based Iterative Learning of the Region of Attraction for Autonomous Systems

This presentation introduces a novel algorithm for estimating the region of attraction of equilibrium points for nonlinear discrete-time autonomous systems. The method iteratively expands an initial estimate of the region of attraction by constructing unions of sublevel sets of learned functions parametrized as neural networks. Unlike conventional techniques that rely on a single global Lyapunov function, the proposed approach provides a collection of local Lyapunov-like functions, enabling richer representations and potentially larger region of attraction estimates. These functions are trained using sampled state-space data, and their Lipschitz continuity ensures that desirable properties extend beyond the training samples. The devised strategy is tested via numerical simulations, demonstrating the effectiveness of the proposed approach.

97 MATHEMATICS AND COMPUTING↗

Attainability analysis in stochastic controlled systems

A control problem for stochastically forced nonlinear continuous-time systems is considered. We propose a method for construction of the regulator that provides a preassigned probabilistic distribution of random states in stochastic equilibrium. Geometric criteria of the controllability are obtained. Constructive technique for the specification of attainability sets is suggested.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Mapping Glacier Basal Sliding Applying Machine Learning

During the RESOLVE project (“High-resolution imaging in subsurface geophysics: development of a multi-instrument platform for interdisciplinary research”), continuous surface displacement and seismic array observations were obtained on Glacier d’Argentière in the French Alps for 35 days in May 2018. The data set is used to perform a detailed study of targeted processes within the highly dynamic cryospheric environment. In particular, the physical processes controlling glacial basal motion are poorly understood and remain challenging to observe directly. Especially in the Alpine region for temperate based glaciers where the ice rapidly responds to changing climatic conditions and thus, processes are strongly intermittent in time and heterogeneous in space. Spatially dense seismic and Global Positioning System (GPS) measurements are analyzed applying machine learning to gain insight into the processes controlling glacial motions of Glacier d’Argentière. Using multiple bandpass-filtered copies of the continuous seismic waveforms, we compute energy-based features, develop a matched field beamforming catalog and include meteorological observations. Features describing the data are analyzed with a gradient boosting decision tree model to directly estimate the GPS displacements from the seismic noise. We posit that features of the seismic noise provide direct access to the dominant parameters that drive displacement on the highly variable and unsteady surface of the glacier. The machine learning model infers daily fluctuations and longer term trends. The results show on-ice displacement rates are strongly modulated by activity at the base of the glacier. The techniques presented provide a new approach to study glacial basal sliding and discover its full complexity.

58 GEOSCIENCES↗

SU-G-JeP1-09: Evaluation of Transperineal Ultrasound Imaging as a Potential Solution for Target Tracking During Ablative Body Radiotherapy for Prostate Cancer

Purpose: Prostate SABR is emerging as a clinically viable, potentially cost effective alternative to prostate IMRT but its adoption is contingent on providing solutions for accurate tracking during beam delivery. Our goal is to evaluate the performance of the Clarity Autoscan ultrasound monitoring system for inter-fractional prostate motion tracking in both phantoms and in-vivo. Methods: In-vivo evaluation was performed under IRB protocol to allow data collection in prostate patients treated with VMAT whereby prostate was imaged through the acoustic window of the perineum. The probe was placed before KV imaging and real-time tracking was started and continued until the end of treatment. Initial absolute 3D positions of fiducials were estimated from KV images. Fiducial positions in MV images subsequently acquired during beam delivery were compared with predicted positions based on Clarity estimated motion. Results: Phantom studies with motion amplitudes of ±1.5, ±3, ±6 mm in lateral direction and ±2 mm in longitudinal direction resulted in tracking errors of −0.03 ± 0.3, −0.04 ± 0.6, −0.2 ± 0.9 mm, respectively, in lateral direction and −0.05 ± 0.30 mm in longitudinal direction. In phantom, measured and predicted fiducial positions in MV images were within 0.1 ± 0.6 mm. Four patients consented to participate in the study and data was acquired over a total of 140 fractions. MV imaging tracking was possible in about 75% of the time (due to occlusion of fiducials) compared to 100% with Clarity. Overall range of estimated motion by Clarity was 0 to 4.0 mm. In-vivo fiducial localization error was 1.2 ± 1.0 mm compared to 1.8 ± 1.9 mm if not taking Clarity estimated motion into account. Conclusion: Real-time transperineal ultrasound tracking reduces uncertainty in prostate position due to intrafractional motion. Research was supported by Elekta.

60 APPLIED LIFE SCIENCES↗